Hi Wahid,
On 18/01/2013 11:37, Wahid Bhimji wrote:
> Hi John
>
> There is some general advice here
> https://www.gridpp.ac.uk/wiki/Performance_and_Tuning
>
> which surprisingly doesn't seem to link to Alessandra's northgrid blog
> on mysql tuning here:
> http://northgrid-tech.blogspot.co.uk/2011/06/dpm-optimization-next-round.html
> (maybe it does and I missed it in a quick scan).
I had found both of these - thanks.
>
> However personally I *don't* think you should drop the requests table
> for this (as that script has got some people in a mess).
> Some of the other mysql tunings may be worth doing - but it doesn't
> sound like you know it is the db that is the problem… (you could log
> slow queries )
> Is there wait Cpu?
I'm getting about 35% wait state.
>
> If you really think that has appeared only because of the change to
> 1.8.5 you could ask the developers on the [log in to unmask]
> <mailto:[log in to unmask]> if something relevant changed.
Thanks, I'll probably do that.
Cheers,
John
>
> Wahid
>
> On 18 Jan 2013, at 11:23, John Hill <[log in to unmask]
> <mailto:[log in to unmask]>> wrote:
>
>> Hi Sam,
>>
>> On 18/01/2013 11:19, Sam Skipsey wrote:
>>> Hi
>>>
>>>
>>> On 18 January 2013 10:44, Mark Slater <[log in to unmask]
>>> <mailto:[log in to unmask]>
>>> <mailto:[log in to unmask]>> wrote:
>>>
>>> Hi John,
>>>
>>> Our SE is not particularly loaded and we have 1.8.5 running. A
>>> couple of things spring to mind:
>>>
>>> 1) Is your SQL DB on the same machine? Ours is split so this may be
>>> a factor....
>>>
>>> 2) Maybe drop the requests tables in the DB? I know this can
>>> certainly have an affect on performance.
>>>
>>> 3) Enable nscd?
>>>
>>>
>>>
>>> Of these, 3) seems the most likely to help in this instance.
>>> I seem to recall that that's what fixed your own issues with newer DPM
>>> installs, Mark? (It also helped out Matt at Lancaster - or the
>>> equivalent process of adding lots of dns entries to the /etc/hosts
>>> config.)
>>
>> As you will have seen, starting nscd hasn't helped me. I might try
>> setting up /etc/hosts manually - though as we have a local DNS server
>> I wouldn't have expected to need to do this.
>> Cheers,
>> John
>>
>>>
>>> Sam
>>>
>>>
>>> I'm guessing others will have more useful ideas but those things
>>> have generally solved our problems in the past!
>>>
>>> Thanks,
>>>
>>> Mark
>>>
>>>
>>>
>>> On 18/01/13 10:27, John Hill wrote:
>>>
>>> Yesterday I upgraded our DPM head node from UMD-1 (DPM 1.8.3) to
>>> EMI-2 (DPM 1.8.5). The upgrade appeared to be successful, but
>>> since then the load average on the SE has stayed stubbornly
>>> above 2 (prior to the upgrade it rarely exceeded 0.2). The SE is
>>> working (all Nagios tests are passing and I can see no obvious
>>> errors in the logs), but some transfers are failing due to
>>> timeouts (so ATLAS opened a ticket overnight).
>>> I tried increasing innodb_buffer_pool_size in /etc/my.cnf and
>>> this may have helped at the 10-20% level. The most active
>>> process is often srmv2.2, but I can't find any discussion of a
>>> need to tune it. Any pointers on how to debug this behaviour
>>> would be welcome.
>>>
>>> John
>>>
>>>
>>
>
>
>
> The University of Edinburgh is a charitable body, registered in
> Scotland, with registration number SC005336.
>
|